There is a lot of baseless declarations about low radar observability in principles and about the Indian version of the PAK in particular, so am going to clear up some misconceptions...
In radar detection, the sphere is consider the ideal body because under radar bombardment, it radiate in a consistent manner. The sphere, of various diameters, is used to calibrate radar systems. With the sphere, we have specular reflection, and depending on the diameter's relationship with the operating frequency, the creeping wave behavior, which is a contributor to the sphere's RCS, may or may not appear.
This relationship is called the 'ten lambda' rule. Lambda is the Greek letter representing frequency or wavelength. If the diameter is greater than 10 times the wavelength (>10), the then creeping wave behavior will not occur. If the diameter is less than 10 times the wavelength (<10), then the creeping wave behavior will occur. So if we are going to calibrate an X-band radar, we should use a sphere whose diameter is >10 frequency in order to get only the specular reflection so we would know as precise as possible the amount of return.
Therefore, the sphere have at most two radiation methods: specular and surface waves.
Going to more complex bodies such as the cube...
The cube have the plate, or more like 6 plates. That is specular reflection. If we tilt a plate, as the angle of tilt increases surface waves behaviors will occur. But when the surface waves reaches a joint between plates, we have a surface discontinuity, the edge, which will give us the radiation mode of diffraction.
So for the left cube, we will have three radiation modes: specular, surface waves, and edge diffraction.
If we introduce curvatures into the cube, like in the right cube, we will still have the same three modes of radiation as the right cube, but less intense for edge diffraction and higher quantity of surface waves.
That means if we rotate both cubes, most likely the left cube will have a higher RCS and pulsating signature than the right cube. With curvatures, we can have better control of how radar signals will leave this finite body than with sharp edges.
So from these three bodies, we, meaning the US, have formulated the basic rules for designing/shaping complex bodies with RCS control in mind.
Control...
- Quantity of radiators
- Array of radiators
- Modes of radiation
The sphere have only its surface for quantity of radiators. The basic cube have two: the plate and edge. The pyramid also have two: the plate and edge. Remember, these are finite bodies, radar signals must leave a body somehow and some time.
On a complex body like an aircraft, major structures like flight control devices are radiators -- the plates. Because they are joined to the fuselage and are in close proximity to each other, this fall under the rule of 'control of array of radiators'. The B-2 have no vertical stabilators. This does not mean the B-2 ignored this rule. On the contrary, the B-2 obeyed this rule perfectly. The rule only say 'control' and part of control is not using a device if there is no need for that device.
When the F-117, F-22, F-35, PAK-FA, and the J-20 canted their vertical stabilators, they also obeyed the rule of 'control of array of radiators'. They canted their vertical stabilators to avoid the dreaded 90-deg corner reflector, which is a major no-no if we want to control final RCS.
The rules for the corner reflectors are:
- Avoid this structure completely.
- If not possible, then avoid the 90-deg type.
The B-2's design obeyed the first rule of the corner reflector structure. The designs for the supposedly 'stealthy' fighters obeyed the second rule of the corner reflector structure.
And both rules for the corner reflector structure falls under the higher rule of 'control of arrays of radiators'.
Once we go beyond this basic understanding, the discussion will involve higher order math and classified data because supercomputers are needed in order to make the complex body of an aircraft as obedient to the top three rules as much as possible.
For something like this...
The only solution is enclosure. A streamline pod that prevents any radar from impacting those structures still obeys those three top rules of RCS control.
So what are we to make of the situation when we have three supposedly 'stealth' fighters that based upon visual inspection, seemingly obey the rules of RCS control to high degree ?
At this point, we have to inspect their bodies at finer granularity. Everything from the degree of curvatures anywhere to panel gaps to fastener heads.
The reason is this...
The above is how any radar system 'sees' an aircraft: as a cluster of voltage spikes created by structures, major and minor, on the aircraft. For the example above, the highest spike came from the tail section, obviously. We can also see a cluster created by the three engines.
One alleged 'stealth' fighter may have its shaping in some ways kept it above the threshold. The J-20 have two ventral fins (control of quantity of radiators) that do not exists on the F-22 and the PAK. Would those structures raise the J-20 above the clutter rejection threshold when they are in the radar stream ? Possibly. We can argue that in the absence of hard data, the J-20 may very well be as 'stealthy' as the F-22, but the first rule of RCS control -- control of
QUANTITY of radiators -- suggests otherwise. The J-20 did not 'violate' this rule. It just did not obey the rule to the same degree as the F-22 does.
The same rules applies to the engine exhausts on all three aircrafts. It is very rare that any radar would have a direct view of the engine exhaust. From this perspective, even the F-22 will be detected. But precisely because it is so rare a situation, the F-22's engine exhausts are designed to be more recessed and better shaped than its competitors so that in the oblique angles, the engine exhausts are less detectable than those on its competitors.
Remember, the rules are:
Control...
- Quantity of radiators
- Array of radiators
- Modes of radiation
We do not 'violate' any rule. We just have varying degrees of
OBEDIENCE to them.
This means the sphere is the near perfect 'stealth' body. This also means that based upon informed opinion, despite the absence of hard data, the J-20 may have a slight edge in terms of better
OBEDIENCE to the RCS control rules than the PAK.